Abstract

SpokenWeb is a research network engaged in the digitization, cataloging and online presentation of audio-visual recordings that document literary activities event held in Canada since the 1950s. Is the network’s goal to develop robust Contents metadata to allow for granular navigation of the audio, and to link this data to other sources, to that end SpokenWeb cataloguers have aimed to add semantic annotations (Wikidata QIDs) to its metadata.

In this presentation we explain the rationale used to develop a web-based tool to help the cataloguers engage in the task of adding semantic annotations to metadata using a semi-automatic approach that ensures high quality annotations while leveraging Natural Language Understanding techniques to accelerate the process.